Hopper-v2

E1274360 UNEXPLORED

Hopper-v2 is a continuous-control reinforcement learning benchmark task in MuJoCo where an agent learns to make a one-legged robot hop forward as efficiently and stably as possible.

All labels observed (1)

Label Occurrences
Hopper-v2 canonical 1

How this entity was disambiguated

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

MuJoCo environments includes Hopper-v2